© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 9852
    Prev
    Next
    INDEX
    Explanations

    destroy or damage objects

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    adamkarvonen/qwen3-32b-saes/saes_Qwen_Qwen3-32B_batch_top_k/resid_post_layer_32
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    èijŃ
    -0.09
    ÑĮев
    -0.09
    overs
    -0.09
     dotyczÄħ
    -0.08
    åķ¬
    -0.08
    ventions
    -0.08
    åĭī强
    -0.08
    hhh
    -0.08
    ways
    -0.08
    yro
    -0.08
    POSITIVE LOGITS
    ively
    0.13
    liness
    0.12
    iveness
    0.11
    inations
    0.11
    æ®Ĩ
    0.10
    สรร
    0.10
    çļ¿
    0.10
    .destroy
    0.10
    destroy
    0.09
    OrUpdate
    0.09
    Activations Density 0.020%

    No Known Activations